Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Sat, 25 Jun 2022 09:17:31 +0200
From:   Lukas Wunner <lukas@wunner.de>
To:     Oleksij Rempel <o.rempel@pengutronix.de>
Cc:     "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>, kernel@pengutronix.de,
        linux-kernel@vger.kernel.org, netdev@vger.kernel.org
Subject: Re: [PATCH net-next v1 1/1] net: phy: ax88772a: fix lost pause
 advertisement configuration
Message-ID: <20220625071731.GA3462@wunner.de>
References: <20220624075558.3141464-1-o.rempel@pengutronix.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220624075558.3141464-1-o.rempel@pengutronix.de>
User-Agent: Mutt/1.10.1 (2018-07-13)
Precedence: bulk

On Fri, Jun 24, 2022 at 09:55:58AM +0200, Oleksij Rempel wrote:
> In case of asix_ax88772a_link_change_notify() workaround, we run soft
> reset which will automatically clear MII_ADVERTISE configuration. The
> PHYlib framework do not know about changed configuration state of the
> PHY, so we need to save and restore all needed configuration registers.
[...]
>  static void asix_ax88772a_link_change_notify(struct phy_device *phydev)
>  {
>  	/* Reset PHY, otherwise MII_LPA will provide outdated information.
>  	 * This issue is reproducible only with some link partner PHYs
>  	 */
> -	if (phydev->state == PHY_NOLINK && phydev->drv->soft_reset)
> +	if (phydev->state == PHY_NOLINK && phydev->drv->soft_reset) {
> +		struct asix_context context;
> +
> +		asix_context_save(phydev, &context);
> +
>  		phydev->drv->soft_reset(phydev);
> +
> +		asix_context_restore(phydev, &context);
> +	}
>  }

Hm, how about just calling phy_init_hw()?  That will perform a
->soft_reset() and also restore the configuration, including
interrupts (which the above does not, but I guess that's
irrelevant as long as the driver uses polling).

Does phy_init_hw() do too much or too little compared to the above
and is hence not a viable solution?

Thanks,

Lukas